Re: mesh: org.bluez.mesh.Element.MessageReceived method does not provide destination address

["Stotland, Inga" <inga.stotland@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 3058 bytes --]

Hi,

On Wed, 2019-09-04 at 13:26 -0700, Gix, Brian wrote:
> On Wed, 2019-09-04 at 21:48 +0200, Michał Lowas-Rzechonek wrote:
> > On 09/04, Michał Lowas-Rzechonek wrote:
> > > The two examples I provided are *not* violating the spec in any
> > > way.
> > > For the record:
> > >  - a combined server/client sitting on element 1 that receives
> > > onoff
> > >    messages and, depending on the destination address, sends a
> > > different
> > >    onoff messages to a "regular" onoff server sitting on element
> > > 0,
> > >    allowing efficient control over switching scenes involving
> > > large
> > >    number of nodes


This sounds like something a vendor model mechanism should handle:
the "mapping" should be understood on both ends: client and server.


> > >  - a model that acts as a IPv6 gateway and directly maps virtual
> > >    addresses to IPv6 addresses of nodes living on the other side
> > > of the
> > >    gateway
> > 
> > Another one about virtual addresses:
> > 
> > In CANOpen, there is a concept of a "Protocol Data Object" [1].
> > Basically, the idea is to pack many pieces of information into a
> > preconfigured format (down to single bits, because CAN frames are
> > even
> > shorter than mesh ones) - this is known as "PDO Mapping Parameters"
> > -
> > then send such payloads to a well-known group address.
> > 
> > In static configurations, this allows to decrease the number (and
> > size)
> > of packets sent by sensor nodes.
> > 
> > Since PDO payloads are *not* self-describing (unlike mesh sensor
> > messages), the receiving party must be aware of the mapping in
> > order to
> > parse the data.
> > 
> > In CANOpen, format is determined by the address - in mesh, it could
> > very
> > well be a virtual label.
> > 
> > [1] https://www.can-cia.org/can-knowledge/canopen/pdo-protocol/
> > 
> 
> I think that this is an interesting use of Virtual Addresses, and in
> addition to this, Mesh Virtual Addresses
> have been suggested as a way of addressing IPv6 addressing as
> well...  However:
> 
> 1. There is already a way something like this could be used
> already:  A model could be created that gets
> subscribed to the Virtual Addresses that require handling by the
> node.
> 
> 2. If such a system proves to be widely requested, daemon support
> could be added (perhaps under a different
> DBus interface) for either or both of IPv6 and "CANOpen".
> 
> In any case the ability to create simple mesh Apps with minimal
> complexity remains intact, and as an added
> bonus, the Open Source community (not to mention the Bluetooth Mesh
> Working Group and larger SIG) can weigh in
> on the preferred methodologies.
> 

My feeling is that the API should be geared towards common case
scenarios (i.e., defined models and such). 

If there is a behavior that absolutely cannot be addressed with the
current API (and use of vendor models), then it has to be
changed/augmented.

As such, I still don't see a compelling reason to do so.



[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3265 bytes --]